Anthropic CEO Dario Amodei proposed pacing AI development via regulation. His plan includes embedded third-party evaluators to verify safety practices. It also calls for democratic coordination to set common safety standards. Global coordination with authoritarian governments is the third element. Amodei argues regulation should target all US frontier AI companies. He acknowledges some coordination may require government support to overcome legal challenges. The proposal aims to balance safety with AI benefits. It follows his essay outlining AI’s risks and potential. The plan risks First Amendment issues in the US. It mirrors existing web content regulation frameworks in other regions.